Survival rate and prognostic factors in advanced cervical cancer patients accompanied by renal impairment

Vol 14, No 3, July – September 2005

Survival servical cancer

173

Survival rate and prognostic factors in advanced cervical cancer patients
accompanied by renal impairment
Imam Rasjidi, Laila Nuranna*, M. Farid Aziz*, Andrijono*, Sigit Purbadi*, Rochani, Nana Supriana§,
Dharmeizar, Bambang Sutrisna #

Abstrak
Tujuan penelitian ini adalah untuk memperoleh informasi tentang survival kanker servik stadium lanjut dengan gangguan fungsi
ginjal (KSSLGFG) serta faktor-faktor yang mempengaruhinya. disamping itu diharapkan dapat ditemukan sistem scoring untuk prediksi kematian pasien KSSLGFG. Rancangan studi yang digunakan adalah retrospektif kohort. Data yang dikumpulkan berasal dari
rekam medik penderita KSSLGFG dari 1 januari 1998 sampai 31 Desember 2003 di Rumah Sakit Umum Pusat Nasional Dr Cipto
Mangunkusumo, Jakarta, dengan jumlah sampel 70 kasus. Hasil penelitian menunjukkan rerata Survival secara keseluruhan pasien
KSSLGFG 8,2 bulan, survival bulan ke-6 sebesar 39%, survival 1 tahun sebesar 3,2%. Median survival 5,3 bulan. Faktor-faktor yang
berpengaruh terhadap survival pasien KSSLGFG adalah gambaran histopatologik (sel adenoskuamosa dan diferensiasi sel), ketebalan korteks ginjal < 1 cm dan nefrostomi. (Med J Indones 2005; 14: 173-8)

Abstract
The objective of this study was to obtain information on the survival rate of advanced cervical cancer patients with renal impairment

(ACCRI) and its prognostic factors. In addition, it is hoped that by this method the scoring system for predicting the death of ACCRI
patients hopetully could be obtained. Design of the study used was retrospective cohort study. Data collected were retrieved from
medical records of ACCRI patients from 1 January 1998 to 31 December 2003 at Dr. Cipto Mangunkusumo National Central General
Hospital, Jakarta, with a total sample of 70 cases. The results of the study showed that mean survival of all ACCRI patients was 8.2 months,
mean survival at sixth month was 39%, and mean survival at one year was 3.2%. Median survival was 5.3 months. Prognostic factors
affecting the survival of ACCRI patients included histopathological type (adenosquamous cell and cell differentiation), cortical thickness
of the kidney less than 1 cm, and nephrostomy. (Med J Indones 2005; 14: 173-8)
Keywords: Advanced cervical cancer, renal impairment, nephrostomy, survival, prognostic factor, scoring system



Fellow of Gynecology oncologic Consultant, Department of
Obstetric and Gynecology, Faculty of Medicine, University of
Indonesia/Dr Cipto Mangunkusumo Hospital, Jakarta,
Indonesia
* Division of Oncology, Department of Obstetrics and
Gynecology, Faculty of Medicine University of Indonesia/
Dr. Cipto Mangunkusumo General Hospital, Jakarta,
Indonesia


Department of Urology, Faculty of Medicine, University of
Indonesia/Dr. Cipto Mangunkusumo Hospital, Jakarta,
Indonesia
§
Department of Radiotherapy, Faculty of Medicine, University of
Indonesia/Dr. Cipto Mangunkusumo Hospital,
Jakarta, Indonesia

Department of Internal Medicine, Faculty of Medicine, University of
Indonesia/Dr. Cipto Mangunkusumo Hospital,
Jakarta, Indonesia
#
Department of Epidemiology, Faculty of Public Health, University of
Indonesia, Jakarta, Indonesia

A total of 250 cases of advanced cervical cancer (62%
of which were stage IIB to IVA)1 were treated at Dr.
Cipto Mangunkusumo General Hospital.
The medical treatment performed on these patients
was radiotherapy. However, most of the patients

showed a low response.
A survey conducted in the United States in 1978
showed that the primary cause of death in patients
with advanced cervical cancer was low response to
radiotherapy or local failure (59%). A study conducted at Dr. Cipto Mangunkusumo General Hospital
1994, on the other hand, revealed that 37 cases (30%
of advanced cervical cancer patients) suffered from
renal failure as a result of ureteral obstruction.3 Most
of the ureteral obstructions in cervical cancers were
associated with mass pressure.

Rasjidi et al

174

This condition made it difficult to overcome the obstruction by cytoscopic retrograde from the bottom.
Renal impairment that also followed made it difficult,
or almost impossible, to perform radiotherapy.
Nephrostomy may become an option to correct renal
impairment, and with it the survival rate of ACCRI

patients.
The aim of this study was to identify the survival rates
and prognostic factors of advanced cervical cancer
patients accompanied by renal impairment. The
present study also attempted to obtain a scoring system for predicting mortality among ACCRI patients.

METHODS
The present study was a retrospective cohort trial,
using data retrieved from medical records of Dr. Cipto
Mangunkusumo General Hospital. All patients with
advanced cervical cancer admitted to Dr. Cipto
Mangunkusumo General Hospital from 1 January
1998 to 31 December 2003 were enrolled. The inclusion
criteria were: 1) Cervical cancer stage IIIB, IVA,
IVB according to FIGO classification4; 2) Recurrent
cervical cancer; 3) Renal impairment approved by the
consultant of Kidney Hypertension Division, Department of Internal Medicine Dr. Cipto Mangunkusumo
National Central General Hospital. The exclusion criteria
were patients lost on follow up.


Med J Indones

teristics: 1) histopathological type; 2) cell differentiation; 3) tumor growth pattern; 4) necrosis; 5) lymphvascular reaction; 6) lymphocyte cells surrounding the
tumor; C) Characteristic factors of renal pathology:
1) hydronephrosis; 2) cortical thickness/ renal condition;
3) non-visualized kidney; 4) hydroureter; D) Factors of
intervention types: 1) conservative; 2) nephrostomy; 3)
type of radiation.
Data were analyzed using SPSS and STATA. We
used Kaplan Meier test to determine survival rate,
while multivariate analysis was used to determine
prognostic factors (Table 3). Scoring system was used
to predict probability of survival after time (Table 4),
and probability of death (Tables 4a and 4b). This
study was approved by ethical committee of Dr. Cipto
Mangunkusumo National Central General Hospital.
We kept the medical records of patients in confidence
in accordance with the applicable medical ethics.

RESULTS

Median survival of six months, and one-year survival
rate of ACCRI patients were 39% and 3.2% , respectively, while median survival was 5.3 months.

Survival Function
1,2

Sample size used the following formula:
2
2
Z
22  Z1  1  2 
 1 2

n
2
1  2 

Where:
1 is rate of group 1 (nephrostomy),
2 is rate of group 2 (Conservative)


1,0

C
U
M
S
U
R
V
I
V
A
L

,8

,6

,4


,2
0,0
-2

0

2

4

6

8

10

12

14


Survival on month

 = ( + ) / 2
1

2

Figure 1. Survival rate of ACCRI patients in month

All patients in the present study were monitored until
their death.
Multivariate analysis
Factors considered to affect survival rate and prognosis in ACCRI patients were as follows: (A) clinical characteristic factors: 1) age; 2) clinical stage; 3)
type of renal impairment; B) histopathological charac-

Multivariate/cox regression analysis showed that risk
factors of survival among ACCRI patients were as
shown in table 1.


Vol 14, No 3, July – September 2005

Survival servical cancer

Table 1. Variables included in making of hazard ratio model
Variable
Age (>= 46 th)

Table 2. Hazard prediction model

HR

95% CI HR

P value

Characteristics

0.73


(0.42;1.25)

0.247

(Non)/ conservative nephrostomy
Histopathological
type: adenocarcinoma
Histopathological
type: adenosquamous
Histopathological
type: small cell
Histopathological
type: unknown
Moderate differentiation
Poor differentiation
Unknown differentiation
Cortical thickness of
right kidney (< 1 cm)

Histopathological type
adenocarcinoma

0.47

(0.14;1.53)

0.209

Adenosquamous

3.10

(1.07;9.00)

0.037

Small cell

0.35

(0.05;2.58)

0.300

Unknown

2.04

(0.94;4.42)

0.071

Differentiation
Moderate

1.15

(0.54;2.45)

0.715

Poor

1.35

(0.53;3.42)

0.533

Unknown

1.95

(0.89;4.28)

0.096

Cortical thickness of
right kidney (< 1 cm)
(Non)/ conservative
nephrostomy
Radiation response
Partial response
Stable
Progressive
Not performed
Blood creatinine
Hemoglobin level

1.72

(0.93;3.20)
0.086

1.78

(1.00;3.16)

1.64
1.53
1.15
2.62
0.98
0.88

(0.55;4.87)
(0.68;3.46)
(0.15;9.10)
(1.25;5.49)
(0.95;1.01)
(0.76;1.02)

0.050
0.373
0.305
0.893
0.011
0.249
0.081

HR

95% CI
HR

Coefficient

z

P
value

Score

2.43

(1.10;5.39)

0.8880

2.18

0.029

20

0.50

(0.11;2.32)

0.6958

0.89

0.376

-8

5.21

(1.61;16.80)

25

(0.02;1.65)

2.76
1.50

0.006

0.20

1.6500
1.6312

0.134

-14

1.41

(0.52;3.83)

0.3411

0.67

0.505

6

1.05
2.28

(0.42;2.67)
(0.76;6.88)

0.0534
0.8245

0.11
1.46

0.910
0.143

1
13

4.25

(1.49;12.15)

1.4474

2.70

0.007

25

2.78

(1.39;5.56)

1.0207

2.88

0.004

26

Table 2 showed final model of multivariate analysis
and it is evident that factors contributing to survival of
ACCRI patients were: nephrostomy (conservative)
was not performed, histopathological type of adenosquamous, unknown cell differentiation, cortical thickness of right kidney < 1 cm.

Table 3. Multivariate analysis of prognostic factors among ACCRI patients
Characteristic
1. Nephrostomy
Yes
No
2. Histophatological subtype
Squamous cell
Adenocarcinoma
Adenosquamous
Small cell
unknown
3. Cell differentiation
Good
Moderate
Poor
Unknown
4. Cortical thickness of right kidney
>= 1 cm
< 1 cm

175

HR

95% CI HR

Coefficient

z

P values

Scores

1
2.43

(1.10;5.39)

0.8880

2.18

0.029

20

1
0.50
5.21
0.20
1.41

(0.11;2.32)
(1.61;16.80)
(0.02;1.65)
(0.52;3.83)

-0.6958
1.6500
-1.6312
0.3411

-0.89
2.76
-1.50
0.67

0.376
0.006
0.134
0.505

-8
25
-14
6

1
1.05
2.28
4.25

(0.42;2.67)
(0.76;6.88)
(1.49;12.15)

0.0534
0.8245
1.4474

0.11
1.46
2.70

0.910
0.143
0.007

1
13
25

1
2.78

(1.39;5.56)

1.0207

2.88

0.004

26

176

Rasjidi et al

Med J Indones

The present study showed that the difference in the
risk or probability between patients who did not undergo
nephrostomy and those who underwent nephrostomy
after being controlled by other variables. It was evident from the results of multivariate analysis that in
the final model, in addition to nephrostomy as a primary variable, there were several confounders, i.e.
histopathological type, cell differentiation, and cortex
thickness of the kidney. Based on the final model, the
probability for patients who did not undergo nephrostomy
to die with the same histopathological type, differentiation cell and renal cortex thickness was 2.43 times
as high as those undergoing nephrostomy.
Histopathological type of adenosquamous. Compared
with squamous type, the mortality risk in ACCRI pa-

tients with adenosquamous cell type was 5.21 times
(HR 5.21, 95%CIHR 1.61-16.80; p = 0.006) with
differetiation cell; intervention and the same cortical
thickness.
Cortical thickness of the kidney. Patients with cortical thickness of right kidney less than 1 cm had a
mortality risk of 2.78 times (HR 2.78; 95% CIHR;
1.39-5.56; p = 0,004) with intervention and histopathology of the Same cell.
In order to predict survival of patients with advanced
stage cervical cancer with renal impairment, we used
formula of survival prediction and probability of
death as follows:

Formula of survival prediction
Survival(t,x)=[S0(t)]
S0(t)
X1
X2
X3
X4

e -(0.8880X1 – 0.6958X2 + 1.6500X3 – 1.6312X4 + 0.3411X5 + 0.0534X6 + 0.8245X7 1.4474X8 1.0207X9)

= Survival base at the –t time
= Conservative( non nephrostomy)
= Histopathological type adenocarcinoma
= Histopathological type adenosquamous
= Histopathological type small cell

X5
X6
X7
X8
X9

= Histopathological type unknown
= Moderate differentiation
= Poor differentiation
= Unknown differentiation
= Cortical thickness of right kidney < 1 cm

Survival probability patients on 365 th day

SACCRI (t,x)=[S0(t)] e -(0,8880X1 - 0,6958X2 + 1,6500X3 - 1,6312X4 + 0,3441X5 + 0,0534X6 + 0,8245X7 + 1,4474X8 + 1,0207X9)
= (0,8139),x =0,01016182=1,02%

survival probability of ACCRI patients on the 365 th day were 1,02%
Probability of death after 365 thday

1
P =

1 + e -(0,8880X1 - 0,6958X2 + 1,6500X3 - 1,6312X4 + 0,3441X5 + 0,0534X6 + 0,8245X7 + 1,4474X8 + 1,0207X9)
p = 1/1012485342
p = 0,98766862
p =98,8 %
Probability of death after 365 days, were 98,8%

Vol 14, No 3, July – September 2005

Survival servical cancer

After calculating the probability of survival, we may also
calculate the scoring system in a clinically practical way.
Score value was the multiplication of individual characteristic (IC) with index value (1). Thus, each case
would have the score that constituted a total of the
above multiplication ( (IC x I)).

177

Table 4b. Death prediction conversion of cervical cancer
Total Scores

Death Prediction

< -7

67 – 70 %

-1 s.d. 11

71 – 80 %

13 s.d. 31

81 – 90 %

> 39

> 90%

Table 4. Enumeration of probability of survival after time (t)
Characteristics
Nephrostomy
Yes
No
Histophatological
subtype
Squamous cell
Adenocarcinoma
Adenosquamous
Small cell
unknown/sensor
Cell differentiation
Good
Moderate
Poor
Unknown/sensor
Cortical thickness of
right kidney
> 1 cm
< 1 cm
Total (NI x K)

Individual
Values
NI

Coefficient

Amount

K

NI x K

0
1

0
0.8880

0.880

0
0
1
0
0

0
-0.6958
1.6500
-1.6312
0.3411

0
0
1

0
0.0534
0.8245
1.4474

0
1

0
1.0207

DISCUSSION
1.6500

0.8245

1.0207
4.3832

Table 4a. Enumeration of probability of survival after time (t)
Characteristics
Nephrostomy
Yes
No
Histophatological
subtype
squamous cell
Adenocarcinoma
Adenosquamous
small cell
Unknown/sensor
Cell differentiation
Good
Moderate
poor
Unknown/sensor
Cortical thickness of
right kidney
> 1 cm
< 1 cm
Total Scores

Individual
Characteristic
KI

Index

Scores

I

KI x I

0
1

0
20

20

0
0
1
0
0

0
-8
25
-14
6

0
0
1

0
1
13
25

0
1

0
26

ACCRI patient did not undergo Nephrostomy (conservative) histopathological type : adenosquamous,
poor differntiation; cortical thicknes < 1 cm Showed
table (4 and 4a) were total score 84.
Since the total score more than 39, death prediction
>90% ( tabel 4b).

25

13

26
84

Probability of survival in ACCRI patients with renal
dysfunction in the 6th month was 39%, while the
cummulative probability in the 12th month was 3.2%,
and median survival was 5.1 months. As shown in
the annual FIGO report, mean survival of cervical
cancer patients was 55%. At the advanced stages, i.e.
stage III and IV, the five-year survival was still high,
i.e. 31.0 % and 7.8 % respectively.
In the present study, with the presence of renal dysfunction the survival rate that the one year survival
drops to 3.2%. The study conducted by Gaspari M
et.al, showed that survival of ACCRI patients was not
influenced by age and renal impairment.9
Harrington KJ et.al: found that the median survival in
all patients with ureteral obstructions leading to malignancies was + 133 days (with a range of 7-7-12).
Seventeen patients (40%) survived up to 6 months
and 5 patients (12%) up to 1 year.8
Median survival after nephrostomy was 526 days, with
a range of 1-712 days. The study of prostate cancer
showed that overall survival rate was 57% for 1 year
and 29% for 2 years (median survival was 21 months).
Another study of prostate cancer found an overall survival rate of 60% for the 1 year and 32% for 2 years.9
With respect to prognostic factors in advanced stage
cervical cancer with renal dysfunction, Culkin DJ et. al
found that the type of histopathology was the main
determining factor for survival in prostate, rectal,
and cervical cancer. However, this was not the case

178

Rasjidi et al

with bladder cancer in which there was no independent factor for significant prognosis.
On the other hand, Roesma J.et.al. found cortical
thickness of < 1 cm, showed that the kidney has suffered an irreversible process. However, nephrostomy
may be considered for cortical thickness > 1 cm.10

Med J Indones

and the staff of Oncology Division, Department of
Obstetrics and Gynecology, Department of Urology,
Department of Radiotherapy, Department of Internal
Medicine, Faculty of Medicine, University of Indonesia,
for their cooperation during this study.

REFERENCES
The aforementioned studies that supported the results
of the present study showed that the prognostic factors
in ACCRI patients were: (1) Type of histopathology
(Adenosquamous cell); (2) Cell differentiation; (3)
cortical thickness (< 1cm), and (4) Nephrostomy.
By taking into consideration the above-mentioned
prognostic factors, we might be able to predict patients’
survivability using survival prediction scoring.

CONCLUSIONS
1. Median survival among ACCRI patients was 5,31
months, while six months and 1 year survival
among ACCRI patients were 39% and 3.2%, respectively. The overall mean of one-year survival of
ACCRI patients was 8.2 months, with survival in
the sixth month at 39%, and survival in one year
at 3.2%. Median survival was 5.31 months.
2. As shown in multivariate analysis, prognostic
factors that affected survival among ACCRI patients included: (a) Histopathological subtype,
(adenosquamous cells and cell differentiation),
(b) cortical thickness